BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Business process management (BPM) is about better managing business processes. This seems self-evident, but many are fixated on the technology and forget the business aspect. Not that technology isn't important -- it certainly is a big part of the devil in which the details lie -- but it is the means to the desired end, not the end itself.
BPM at the center
SOA, BPM and modeling are not three routes to improved processes; rather, they are three threads that can be woven into a highly functional process fabric. A look at their basic functionality reveals why:
- BPM can be characterized as the utility that uses rules and a runtime engine to manage and execute business processes.
- Modeling can be described as the blueprint that illustrates how -- and how well -- various processes flow.
- SOA can be part of the architectural underpinning that enables and supports the orchestration of the computing power needed for BPM.
As you can see, SOA, BPM and modeling each serve a specific purpose, but they are not all indispensable to the task. Rather, BPM is the main thread around which the other two can be braided because it is home to the basic capabilities of business rules, workflows, process analytics, etc. -- namely, the things that make BPM, BPM. Believe it or not, BPM can be done without either modeling or SOA.
However, including these two elements certainly can make BPM more effective. By utilizing deep modeling tools, one can document and simulate process flows, plus layer on and manipulate infrastructure assumptions so the performance impact of adding a server here, installing a CPU there, or leveraging SOA to increase a system's speed or flexibility can be seen.
On the other hand, SOA's strength lies in its ability to support communications between services by defining how two computing entities interact to enable one entity to perform a unit of work on behalf of another. In BPM terms, this means the rules repository can live behind a firewall, the execution engine in the cloud, and the modeling capability off-site with the business analyst consultant hired to balance the process loading and ensure performance objectives are met.
Connections between SOA, BPM are key
In many ways, discussions about SOA and BPM are really conversations about connective tissue. Are the various processing locations connected via an API? Web service wrappers? Are they hard-wired within a single server? Where are they located? On-premises? In the cloud?
These questions need to be informed by operational requirements. One intensifying need these days involves embracing mobile participants in organizational workflows, an imperative that adds different sorts of demands to connectivity and security, and may contribute to a decision to adopt SOA.
This is why it's important to look beyond the technology and fully consider how each one can bring out the practical best in the others. For in the end, it's all about improving the management of business processes and not determining which technology is better. This means evaluating how best to weave different technology threads into a single process fabric.
About the author:
Steve Weissman has a 20-year track record of innovation and success in helping organizations derive maximum total value from their information solutions. A seasoned consultant, analyst, and professional trainer, he uses his keen strategy, business, and technology skills to identify, measure, and mesh his clients’ needs and goals, and recommend effective best-practices and solutions for managing processes, content, and data. He can be reached at email@example.com or 617-383-4655.
How SOA, BPM are fueling interest in data modeling
Why SOA and BPM are increasingly being used together
How combining BPM and SOA can improve processes